Benchmark result
gemma-4-31B-it-qat-q4_0-uncensored-heretic-NVFP4 on 3× NVIDIA GeForce RTX 5090 — 55.2 tok/s
Measured with llama.cpp b10590-6657ded4f on September 4, 2026.
What the model built
Open full screen →The coding scenario asks for a playable game in a single HTML file. This is exactly what the model returned, unedited; the frame only adds a line that tells this page how tall it is.
Reproduce this run
toolllama.cpp b10590-6657ded4f
modelgemma-4-31B-it-qat-q4_0-uncensored-heretic-NVFP4 (Q4_0)
context8,192
samplingtemperature 1 · top_p 0.949999988079071 · top_k 64 · min_p 0.05000000074505806 · presence_penalty 0 · repeat_penalty 1
clientv0.4.55+97
Set those in llama.cpp, then:
llm-benchmark benchmark --model "gemma-4-31B-it-qat-q4_0-uncensored-heretic-NVFP4" --tool "llama.cpp"The client prompts for context length and thinking mode, and for the KV cache dtype on Unsloth Studio. Sampling is left at the model default — the values above are what the tool reported using, not overrides the client sent.